Risk-adjusted predictive bidding for electronic advertisements

1. A computerized method for risk-adjusted predictive bidding for electronic advertisements during loading of a webpage, the method comprising:

receiving, by a first computing device from a software application of a requesting device during loading of a webpage, a first request for graphical display source code corresponding to a computerized graphical advertisement display to be inserted into one or more impression opportunities on the webpage;

transmitting, by a second computing device coupled to the first computing device, during loading of the webpage by the requesting device, a second request to a plurality of third-party computing devices for one or more bids for an impression opportunity of the one or more impression opportunities;

receiving, by the second computing device, during loading of the webpage by the requesting device, the one or more bids from the plurality of third-party computing devices;

adjusting, by the second computing device, during loading of the webpage by the requesting device, at least one of the one or more bids based upon a risk tolerance parameter (α) received from the third-party computing device that submitted the corresponding bid, wherein the risk tolerance parameter (α) as received is encoded as a key-value pair in a message returned from the third-party computing devices in response to an application programming interface (API) call initiated by the second computing device, and including:

applying the risk tolerance parameter (α) received from the third-party computing device to a bid adjustment algorithm, wherein the risk tolerance parameter (α) modifies a standard deviation value associated with a probability of a sale given an interaction with the impression opportunity by a user of the webpage,

wherein the bid adjustment algorithm is

CPA×p (click)× p (sale|click)

where CPA is cost per action, p(click) is a probability of a click on the advertising impression, and p(sale|click) is a probability of a sale given a click on the advertising impression, and

wherein the risk tolerance parameter (α) is applied to the bid adjustment algorithm as:

CPA×p (click)× p (sale|click)×(1−α×stdev( p (sale|click)))

where stdev(p(sale|click)) is the standard deviation of the probability of a sale given a click on the advertising impression;

determining, by the second computing device, during loading of the webpage by the requesting device, whether to select the impression opportunity based on the adjusted one or more bids; and

generating, by the first computing device, during loading of the webpage by the requesting device, graphical display source code based upon a bid of the adjusted one or more bids for display on the requesting device in the webpage upon completion of loading of the webpage.

2. The method of claim 1 , wherein the adjusting step comprises lowering the bid based upon the risk factor.

3. The method of claim 1 , wherein the risk tolerance parameter (α) represents a risk that the computerized graphical advertisement display inserted into the impression opportunity is not interacted with by a user of the webpage.

4. The method of claim 1 , wherein the interaction with the computerized graphical advertisement display comprises a click of the computerized graphical advertisement display, or a view associated with the computerized graphical advertisement display.

5. The method of claim 1 , wherein the risk tolerance parameter (α) is based upon an expected long-term value to an advertiser associated with the third-party computing device that submitted the corresponding bid.

6. The method of claim 1 , wherein the risk tolerance parameter (α) is based upon a risk preference indicator received by the second computing device from the third-party computing device as part of the corresponding bid.

7. The method of claim 6 , wherein the risk preference indicator identifies whether an advertiser associated with the third-party computing device is risk-sensitive.

8. The method of claim 1 , further comprising:

when the impression opportunity is selected:

storing, by the first computing device, advertisement information related to the impression opportunity;

generating, by the first computing device, graphical display source code for a computerized graphical advertisement display based upon the advertisement information and indicia of an advertiser associated with the third-party computing device that submitted the selected bid; and

providing, by the first computing device, the graphical display source code to the requesting device.

9. The method of claim 8 , further comprising:

rendering, by the requesting device, the graphical display source code to display the computerized graphical advertisement display.

10. The method of claim 1 , further comprising:

determining, by the second computing device, a value associated with selecting the impression opportunity;

determining, by the second computing device, a largest bid of the adjusted one or more bids; and

selecting, by the second computing device, the impression opportunity using the largest bid if the value associated with selecting the impression opportunity is greater than or equal to the cost associated with selecting the impression opportunity.